Sensitivity analysis in decision circuits

TL;DR

This paper explores how decision circuits can be used for sensitivity analysis in influence diagrams, enabling efficient evaluation of how changes in model parameters affect optimal strategies and information value.

Contribution

It introduces a method leveraging decision circuits' derivative information for sensitivity analysis in sequential decision problems.

Findings

Efficient computation of value of information.

Analysis of how model parameter changes impact optimal strategies.

Utilization of derivative information in decision circuits.

Abstract

Decision circuits have been developed to perform efficient evaluation of influence diagrams [Bhattacharjya and Shachter, 2007], building on the advances in arithmetic circuits for belief network inference [Darwiche,2003]. In the process of model building and analysis, we perform sensitivity analysis to understand how the optimal solution changes in response to changes in the model. When sequential decision problems under uncertainty are represented as decision circuits, we can exploit the efficient solution process embodied in the decision circuit and the wealth of derivative information available to compute the value of information for the uncertainties in the problem and the effects of changes to model parameters on the value and the optimal strategy.